https://orcid.org/0000-0001-5967-864X
Abstract
Despite the recent advent of molecular data to assess the phylogeny of many invertebrate groups, the systematics of the pseudoscorpion family Chernetidae is unresolved, even though it comprises a quarter of the world’s generic pseudoscorpion diversity. We derive a preliminary molecular phylogeny of chernetids to assess subfamilial and generic monophyly using mitochondrial (COI) and nuclear (28S rRNA, 18S rRNA, and H3) markers. Three chernetid subfamilies have previously been recognized and Lamprochernetinae, originally defined based on T-shaped spermathecae, is recovered here but expanded to include the Old-World tropical genus Verrucachernes. In contrast, the genera Conicochernes and Calymmachernes of the subfamily Goniochernetinae are nested within the largest subfamily, Chernetinae. Three new subterranean species of the Palearctic genus Lasiochernes are also described from the Dinaric Karst: L. marinaeHlebec & Harvey, sp. nov., L. jalziciHlebec & Harvey, sp. nov., and L. pavlekaeHlebec & Harvey, sp. nov.. The former two species are single-cave endemics but L. pavlekae sp. nov. is more widespread. The relative lack of genetic structuring in this species, including haplotype sharing between sampling localities, together with a patchy distribution, suggests that its biogeography is probably shaped by multiple vector-mediated dispersal events, rather than geomorphological history. Due to their rarity, we discourage further collecting of Lasiochernes in this subterranean biodiversity hotspot.
Introduction
The Dinaric Karst of the Western Balkans stands out as the world’s most prominent subterranean biodiversity hotspot and harbours a plethora of remarkable specialized species (Gottstein Matočec et al. 2002), including the first scientifically described cave animal, the aquatic salamander Proteus anguinus Laurenti, 1768 (Laurenti 1768). Not surprisingly, paleogeographic and historical climatic events in the wider Mediterranean area, coupled with ecological speciation, have played a key role in shaping the natural histories of these unique subterranean taxa (Sket 1999). Among the least comprehensively explored taxa are pseudoscorpions (Arachnida: Pseudoscorpiones), one of the most diverse invertebrate lineages in the Dinaric Karst (Hlebec et al. 2023), although research commenced in the mid-19th century (Schiödte 1847, Schmidt 1848) and 143 species have been described from this area (e.g. Beier 1939, Ćurčić 1988, WPC 2022). Most of these species are strongly troglobitic and lack eyes or have reduced body pigmentation and attenuated appendages. The majority of the species from Croatia belong to the families Chthoniidae (43 species) and Neobisiidae (79 species). These families are taxonomically complex and require systematic revision. It is, therefore, unfortunate that most Dinaric species have only been described based on morphology and small samples, without an attempt to elucidate intraspecific variability in phenotypic traits or biogeographical patterns. This poses a problem due to the cryptic nature of pseudoscorpion species, potential homoplasy in morphological characters, and high alpha-level diversity in the area.
The family Chernetidae is one of the three most speciose pseudoscorpion families, encompassing over 690 species across 118 genera (WPC 2022). In contrast to the Chthoniidae and Neobisiidae, the family is also ecologically diverse and includes species with complex mating behaviours, phoresy as a means of dispersal, and diverse habitat choices that even include animal dung and hay (Finlayson et al. 2015, Harvey et al. 2015, Nassirkhani 2016, Tapia-Ramírez et al. 2022). The European species, in particular, have been relatively well studied, although putatively cryptic taxa with significant morphological stasis have been reported (Muster et al. 2021, Christophoryová et al. 2023). Recent studies are providing valuable information on the geographic distribution of individual species (e.g. Opatova and Št’áhlavský 2018), which is often a consequence of phoretic associations (Beier 1948) with flying insects (Červená et al. 2019). Taxonomic revisions, biogeographic patterns, and phylogenetic relationships of chernetids are usually neglected and the group is in a rather chaotic taxonomic state. It is, therefore, unsurprising that the three recognized subfamilies are diagnosed poorly. Goniochernetinae Beier, 1932 (three genera, 10 species) are characterized by the presence of an angulate posterior margin of the carapace; Lamprochernetinae Beier, 1932 (eight genera, 117 species) are defined by the presence of T-shaped spermathecae in females; and Chernetinae Menge, 1855 serve as a ‘basket’ taxon lacking these specific character states (Harvey 1995, Harvey et al. 2012). However, very few of the 118 named genera have ever been sequenced, so published phylogenies that include chernetid taxa (Murienne et al. 2008) are limited in answering questions about their monophyly and classification.
The subfamily Lamprochernetinae currently contains eight genera, most of which occur in the Holarctic region, although some genera are found in tropical regions. The genus Lasiochernes Beier, 1932 (10 described species) is easily recognized as Lamprochernetinae due to the presence of T-shaped spermathecae in females (Harvey 1994). Nine species of the genus are range-restricted and rarely collected (Christophoryová et al. 2016), while only Lasiochernes pilosus Ellingsen, 1910 has a broader distribution in Europe (WPC 2022), presumably due to its close association with moles (Talpa spp.). Each species displays significant sexual dimorphism, with males typically bearing long setae on the pedipalpal segments, especially the femur. Most species are known only from isolated species descriptions, but a recent study (Christophoryová et al. 2016) employed multivariate morphometrics to assess differences between Lasiochernes cretonatus Henderickx, 1998, L. jonicus Beier, 1929, and L. pilosus, documenting variation in many frequently used taxonomic characters.
Recent collecting in caves in the southern part of the Dinaric Karst in Croatia has led to the discovery of three morphologically and genetically distinct lineages of Lasiochernes, thus extending the known range of this genus within the Mediterranean region. In this study, we analyse morphological and molecular data (both nuclear and mitochondrial) from 29 species of Chernetidae to achieve the following objectives: (i) test the monophyly of the subfamilies Lamprochernetinae, Goniochernetinae, and Chernetinae; (ii) provide the first molecular data for Lasiochernes and employ species delimitation approaches to confirm the presence of evolutionary distinct units in Croatia; (iii) describe three new species and present a diagnostic key for all currently recognized species within the genus; and (iv) re-evaluate the significance of morphological traits used for differentiation.
Methods
Taxon sampling and identification
To assess the monophyly of the subfamilies Lamprochernetinae, Goniochernetinae, and Chernetinae (Fig. 1), we conducted phylogenetic analyses using representatives of 21 chernetid genera, including the type species of 10 genera. We incorporated newly obtained chernetid sequences (Supporting Information, Appendix S1), along with sequences of other chernetids downloaded from GenBank (for the list of sequences and GenBank accession numbers, refer to Supporting Information, Appendix S2). Each species was represented by a single terminal, except Dinocheirus panzeri (C. L. Koch, 1836), Verrucachernes sp. Chamberlin, 1947, Lamprochernes nodosus (Schrank, 1803), Lamprochernes savignyi (Simon, 1881), and Allochernes cf. balcanicus Hadži, 1938, for which we included two or more individuals, aiming to assess intraspecific genetic divergences.
Cladograms showing the hypothesized relationships of chernetid subfamilies and genera. A, classification discussed by Harvey (1995). B, classification based on the phylogenetic analysis. Bold font indicates genera sequenced within the present study.
We collected 47 specimens of Lasiochernes from 20 caves and pits locally referred to as ‘špilja’ and ‘jama’, respectively (Fig. 2). Specimens were preserved in 96% alcohol and were studied using temporary slide mounts prepared by the dissection of the chela, chelicera, and leg IV from the body, immersion in lactic acid for several days while mounted on microscope slides. After examination, the specimens were rinsed in water and returned to 75% ethanol, with the dissected portions placed in glass genitalia microvials (BioQuip Products, Inc., Compton, CA, USA). Specimen examination was carried out using Olympus BH–2 compound and Leica MZ–16A dissecting microscopes, and digital images were captured using Leica Application Suite v.3.7.0 software. Line drawings were prepared using a Leica drawing tube attached to an Olympus BH–2 compound microscope. Additionally, two specimens of the putative new species, Lasiochernes pavlekae sp. nov., were critically point-dried, gold-coated, and imaged using a Hitachi TM3030 Plus scanning electron microscope. Terminology follows Chamberlin (1931), Harvey (1992a) for modifications to the trichobothria terminology, and Judson (2007) for the chelicera. Abbreviations for trichobotria include: eb, external basal; esb, external sub-basal; ib, internal basal; isb, internal sub-basal; ist, internal sub-terminal; est, external sub-terminal; it, internal terminal; et, external terminal; t, terminal; b, basal; sb, sub-basal; st, sub-terminal. Measurements were taken at the highest magnification using an ocular graticule, and all figures were assembled in Adobe Illustrator. The specimens used in the descriptions of new species are lodged in the following collections: Croatian Biospeleological Society Collection (CBSS), Collection of Scorpiones and Pseudoscorpiones at the Croatian Natural History Museum (CNHM), and the Western Australian Museum (WAM).
Geographic distribution of Lasiochernes species. The main map (B) is an enlarged area marked in the map (A). The red line (in A) indicates the Dinaric Karst. Collecting sites with black circles (in B) are type localities of the species. C, life habitus of Lasiochernes marinae sp. nov.; D, life habitus of Lasiochernes pavlekae sp. nov.. Photos are courtesy of Branko Jalžić.
Molecular methods and alignment
This study aims to characterize several genera of the family Chernetidae, including Haplochernes Beier, 1932, Barbaraella Harvey, 1995, Austrochernes Beier, 1932, Balgachernes Harvey, 2018, Nesidiochernes Beier, 1957, Sundochernes Beier, 1932, Verrucachernes Chamberlin, 1947, Megachernes Beier, 1932, and Lasiochernes based on standard molecular markers. Genomic DNA was extracted from 20 specimens using one leg and either the QIAamp DNA Micro Kit (Qiagen, Hilden, Germany) or the DNeasy blood and tissue kits (Qiagen, Melbourne, Australia), following the manufacturer’s protocol. However, 50 µL of elution buffer was used to increase DNA yield. Partial fragments of four genes: one mitochondrial-barcode region of the cytochrome c oxidase subunit I (hereafter ‘COI’) and three nuclear-18S rRNA (hereafter ‘18S’), the domain I region of the 28S rRNA (hereafter ‘28S’), and Histone H3 (hereafter ‘H3’) were sequenced using the primers and protocols listed in the Supporting Information, Appendix S3. These genes have been routinely used in phylogenetic analyses of various pseudoscorpion taxa and allowed for comparisons with previously published sequences. Purification and bi-directional sequencing of polymerase chain reaction (PCR) products using amplification primers was done by Macrogen Inc. (Amsterdam, The Netherlands) and the Australian Genome Research Facility (AGRF).
Sequence data authentication and phylogenetic analyses
Sequence chromatograms were checked for ambiguities and double peaks, manually edited, and assembled in Geneious Prime 2022.1 (Biomatters, Auckland, New Zealand). Amino acid translations and checks for stop codons were conducted using MESQUITE v.3.61 (Maddison and Maddison 2008). The newly generated sequences were deposited in GenBank (accession numbers available in Supporting Information, Appendix S2) and the BOLD database (under the dataset Lasiochernes, DOI: https://doi.org/10.5883/DS-LASIOCH). The gene fragments were aligned separately using MAFFT v.7 (Katoh et al. 2019) with the ‘auto’ strategy for the COI and H3 datasets. For the 28S and 18S datasets, the Q-INS-i algorithm (Katoh and Toh 2008), which considers the secondary structure of RNA, with default settings, was employed. As the ribosomal gene alignments did not contain long gaps or ambiguously aligned regions, the entire alignments were used to generate a concatenated alignment in Geneious Prime 2022.1. Phylogenetic analyses were conducted using maximum likelihood (ML) (Felsenstein 1973) and Bayesian inference (BI) methods for the complete data matrix, as well as for each locus separately. The best partitioning scheme and substitution models were determined using PartitionFinder v.2.1.1 (Lanfear et al. 2017) based on the Bayesian information criterion (BIC) and the greedy algorithm. The complete dataset was partitioned by gene and codon positions (for the two protein-coding genes). The best-fit partitioning schemes and nucleotide substitution models are listed in Supporting Information, Appendix S4. ML phylogenetic analyses were performed using IQ-TREE v.2.0.3 (Minh et al. 2020), and the support for internal nodes was assessed using 5000 ultrafast bootstrap replicates (UFBoot) (Hoang et al. 2018). Neobisium geronense Beier, 1939 (JN018184) and Chthonius ischnocheles (Hermann, 1804) (JN018172) were used as outgroups, and the datasets were rooted against the scorpion Euscorpius italicus (Herbst, 1800) (MK421715). BI analysis was conducted using MrBayes v.3.2.7 (Ronquist et al. 2012) with the optimal substitution models determined by PartitionFinder v.2.1.1. Two independent Markov Chain Monte Carlo (MCMC) algorithms with four chains each, were run simultaneously for 50 million generations, with sampling of every 1000th tree. The resulting log-files were analysed using TRACER v.1.7.1 (Rambaut et al. 2018), and the first 20% of sampled trees were discarded as burn-in. The remaining trees were assembled to create a majority-rule consensus tree, and Bayesian posterior probabilities (BPP) were used to assess node confidence. Phylogenetic trees were annotated using FigTree v.1.4.3 (available from https://tree.bio.ed.ac.uk/software/figtree) and iTOL v.5 (Letunic and Bork 2021). All analyses were performed using the CIPRES Science Gateway v.3.3 (Miller et al. 2010).
Species delimitation and nucleotide divergence
The taxonomic status of pseudoscorpions described from Croatia is far from being satisfactory, with many poorly described species based on single specimens or lacking precise locality data, and absence of specimens in public museum collections. To improve from here, we employed three approaches (two distance-based and one tree-based) to assess the agreement between clustering of COI sequences and morphology-based delineation before describing each of the new species. We applied the following methods: automatic barcode gap discovery (ABGD) (Puillandre et al. 2012), assemble species by automatic partitioning (ASAP) (Puillandre et al. 2020), and the Bayesian implementation of the Poisson tree process (bPTP) (Zhang et al. 2013). These analyses were run on the respective servers (https://bioinfo.mnhn.fr/abi/public/abgd; https://bioinfo.mnhn.fr/abi/public/asap) using default settings, except for the ABGD analysis where X value (relative gap width) had to be lowered to 1 since the default settings returned only one partition. For ASAP, we applied a p-distance model and set the recursive split probability to 0.01. In the bPTP analysis, the BI COI tree was used as the input tree, excluding the outgroups (Zhang et al. 2013). The analysis was run on the web-server (https://species.h-its.org/ptp). Bayesian posterior probabilities for putative species were acquired after running 500 000 MCMC generations, thinning the set to 100, sampling every 500 generations, and discarding the first 20% of samples as burn-in. To assess intra- and interspecific genetic diversity, uncorrected pairwise distances (p-distances) for all gene fragments were calculated in MEGA-X v.10.2.6 (Kumar et al. 2018), treating missing data with the pairwise deletion option. Furthermore, to examine COI haplotype diversity and relationships at the intraspecific level for Lasiochernes pavlekae sp. nov., a median-joining network (Bandelt et al. 1999) was generated using PopART v.1.7 (Leigh and Bryant 2015).
Morphometry and statistics
In order to detect potential intra- and interpopulation morphological variations, and to test for morphometric clusters, we examined specimens from two thoroughly sampled populations: the ‘island’ (Mljet) population and the ‘mainland’ (Pelješac) population, tentatively identified as Lasiochernes pavlekae sp. nov.. Specimens were temporarily mounted on microscope slides in lactic acid and examined under an Olympus BH–2 compound microscope. We measured 11 traits (refer to the Supporting Information, Appendix S5 for the list of characters and raw morphometric measurements) for a total of 13 males and 16 females, and analysed them using a principal component analysis (PCA). All measurements were taken at the highest magnification using an ocular graticule. The selected characters were deemed taxonomically relevant based on previous studies on Lasiochernes, including the length of the cheliceral movable finger that is a reliable diagnostic character in females (Christophoryová et al. 2016). To account for different scales of measured traits, all the measurements were scaled to zero mean and unit variance. The PCA was conducted in PYTHON v.3.11.1 using SCIKIT-LEARN v.1.2.1 (Pedregosa et al. 2011), and the results were visualized using MATPLOTLIB v.3.6.3 (Hunter 2007). Student’s t-test was used to assess differences between the two populations for traits that met the assumptions of normality (tested with the Shapiro–Wilk test) and homogeneity of variance (tested with Levene’s test). For traits that did not meet these assumptions, a Mann–Whitney U rank test was employed.
Results
Sequence data
The final alignments were 658 bp long for COI, 1223 bp for 28S, 1781 bp for 18S, and 334 bp for H3. Some taxa had only two or three available fragments (refer to the Supporting Information, Appendix S2 for detailed information). The concatenated matrix used for analyses included a total of 47 terminals, consisting of 44 chernetids (29 species) and three outgroups, with a total of 3996 aligned characters. Among these, 1715 sites were variable.
Molecular phylogeny
The molecular phylogenetic analyses using ML and BI on the concatenated dataset produced congruent and well-resolved trees. The majority of clades showed strong statistical support (Fig. 3), and the overall topology agreed with the single-locus phylogenies.
Phylogram derived from ML analysis of concatenated dataset (47 terminals, four genes, 3996 bp) with species delineation results and spermathecae shape. Morphospecies were mapped onto the terminals of the tree along with COI GenBank accession numbers. Branch support is given as ultrafast bootstrap replicates (UFBoot)/Bayesian posterior probabilities (BPP). Colour strips refer to species delimitation results (MOTUs) as indicated by ABGD, ASAP, and bPTP. Spermathecae drawings from original descriptions and redescriptions (Mahnert 1978, 2011, Harvey 1988, 1992b, 1995, 2018, Christophoryová et al. 2012, Nassirkhani et al. 2015, Nassirkhani and Shoushtari 2016, 2017, Gao et al. 2017, Nassirkhani 2018, Gao and Zhang 2020). All scale lines = 0.1 mm, except for Balgachernes spermathecae (scale line = 0.05 mm).
The genera that are traditionally included in the subfamily Lamprochernetinae (Lamprochernes Tömösváry, 1883, Allochernes Beier, 1932, Pselaphochernes Beier, 1932, Megachernes, and Lasiochernes) form a clade that is highly supported (ML UFBoot = 100 and BPP = 1.00). As previously discussed by Harvey (1995), this subfamily is morphologically diagnosed by the presence of T-shaped spermathecae. Interestingly, the cosmotropical genus Verrucachernes also clusters within this clade, supported by strong molecular evidence (ML UFBoot = 99 and BPP = 1.00) and similarities in female genital morphology. Although the spermathecae of Verrucachernes are not T-shaped, they comprise a single median duct and a terminal spermathecal receptaculum (Chamberlin 1947, Harvey 1988, Romero-Ortiz and Harvey 2019), which seems to represent a precursor state to the T-shaped morphology. Therefore, the diagnosis for the subfamily Lamprochernetinae should now refer to the presence of a single median duct, rather than the paired ducts found in other chernetids. Additionally, the analysis strongly recovers Orochernes ganziensis Gao and Zhang, 2019 as a sister-lineage to the Lamprochernetinae (ML UFBoot = 100 and BPP = 1.00). However, we refrain from including Orochernes Beier, 1968 in this subfamily due to the paired spermathecae (Gao and Zhang 2019).
Within the subfamily Lamprochernetinae, the analysis recovered Verrucachernes as sister to the other genera, followed by Lamprochernes sister to a clade comprising Allochernes, Pselaphochernes, Megachernes, and Lasiochernes. The latter two genera share a similar habitus and seem to occur mostly in association with mammals or mammal guano in caves (e.g. Henderickx 1998, Harvey et al. 2012, Okabe et al. 2020), suggesting a similar ecological niche. Both genera are largely allopatric with Lasiochernes occurring in south-east Europe, Middle East, and Central Africa, and Megachernes occurring west of the Black Sea to East Asia and south to Australia (WPC 2022). The molecular evidence (Fig. 3), as well as morphological data (e.g. enlarged and rounded posterolateral corner of coxa IV and presence of very long female spermathecae restricted to Megachernes), support the distinction between these two genera, with Pselaphochernes (ML UFBoot = 100 and BPP = 1.00) identified as a sister-taxon in the phylogeny. Three newly described Lasiochernes species, discovered in Croatian caves, form a well-supported distinct clade (Fig. 3). Two of these species, Lasiochernes jalzici sp. nov. and Lasiochernes marinae sp. nov. appear as sister-taxa (ML UFBoot = 86 and BPP = 0.99).
The recovered phylogeny did not match current subfamily structure. While the subfamily Lamprochernetinae was recovered as monophyletic, the subfamily Goniochernetinae is nested within the larger subfamily Chernetinae (Figs 1, 3). Chernetinae, in turn, is a diverse group for which diagnostic features have yet to be determined. The genus Haplochernes appeared as the most basal branch among the chernetids. Chernetinae includes a range of genera and is sister to the Orochernes + Lamprochernetinae clade. Within Chernetinae, a well-defined ‘Chernes-clade’ (ML UFBoot = 100 and BPP = 1.00) was observed, consisting of Chernes, Dendrochernes Beier, 1932 and Dinocheirus Chamberlin, 1929. These genera, along with Hesperochernes Chamberlin, 1924, possess paired, long, slender spermathecae (Muchmore 1974) and all occur in the Northern Hemisphere. Additionally, there is a strongly supported ‘Australian-clade’ (Austrochernes Beier, 1932, Conicochernes Beier 1948, Calymmachernes Beier, 1954, Marachernes Harvey, 1992, Balgachernes, Nesidiochernes, and Sundochernes) (ML UFBoot = 100 and BPP = 1.00), but we are not aware of any morphological synapomorphy (Harvey, unpublished data). Two genera currently assigned to the Goniochernetinae (three genera, 10 species): Conicochernes and Calymmachernes are nested within Chernetinae. We conclude that the status of this subfamily is still not resolved, especially as we have not been able to include the type genus Goniochernes Beier, 1932.
We also note that the phylogenetic positions of Orochernes ganziensis and Haplochernes boncicus Karsch, 1881 require re-testing in the future. These species appeared as distinct genetic lineages not placed within any of the currently recognized subfamilies (Fig. 3).
Species delineation-COI analysis
Molecular species delimitation results are consistent with the morphology-based species hypotheses. Through an integrative approach, we have identified three species: Lasiochernes marinae sp. nov., Lasiochernes jalzici sp. nov., and Lasiochernes pavlekae sp. nov. Their diagnoses are provided below. See the Supporting Information, Appendix S6 for a list of diagnostic nucleotides in the barcode fragment of the COI gene.
The ABGD analysis delimits 31 putative species based on both Jukes–Cantor (JC69) and Kimura (K80) models. Similarly, the ASAP analysis, with the best ASAP-score 6.50, infers the same number of putative species. Both methods distinguish the specimen from Špilja Crno jezero cave (GenBank OQ330000) from another L. pavlekae sp. nov. specimens. On the other hand, the bPTP method indicates the highest number of putative species (36), further dividing Lasiochernes pavlekae sp. nov. into seven groups according to collection sites. However, four of these groups have low confidence support (BS value < 0.50) and exhibit incongruence with the morphology-based findings (see details below).
Diversity of Lasiochernes pavlekae
Based on our extensive sampling efforts, L. pavlekae sp. nov. exhibits low haplotype diversity, and haplotypes from different localities are separated by a small number of mutational steps. Haplotypes are shared by individuals from three areas: Vilina špilja-Izvor Omble cave, Pelješac Peninsula, and Mljet Island, as well as between Mljet Island and Gustac Islet in the Kornati archipelago. However, the specimen from Špilja Crno jezero cave (GenBank OQ330000) represents a genetically distinct population, separated by seven mutational steps from the central haplotype (Fig. 14A).
The principal component analysis (PCA) identified the traits that account for the majority of the morphological variance. The first three principal components explain a total of 76% of the variances. As outlined in Supporting Information, Appendix S7, PC1 (39.9% variance) strongly correlates with chela and hand ratios (with and without pedicel), PC2 (20.8% variance) correlates with body length, carapace length, and width, while PC3 (15.3% variance) correlates with ratios of the palp segments (trochanter, femur, and patella ratios). This indicates that the taxonomic signal primarily arises from measurements derived from body appendages and size ratios. Among all measured traits, only the palpal movable finger ratio does not show a correlation with the first three PC axes, and thus, this character was excluded from further analyses. The results of Student’s t-test and Mann–Whitney U rank test reveal no statistically significant differences in the analysed traits between the ‘island’ (Mljet) and ‘mainland’ (Pelješac) populations. Furthermore, the overlap of morphometric ranges of measured traits between the two populations on first two PC axes suggests that all specimens belong to a single species (Fig. 14C). In contrast, a significant morphological differentiation between males and females is evident (Fig. 14D), which is expected considering the pronounced sexual dimorphism in Lasiochernes. Please refer to the Supporting Information, Appendix S7 for the full matrix of PCA components and Appendix S8 for the statistical analysis.
Species descriptions
Order Pseudoscorpiones De Geer, 1778
Family Chernetidae Menge, 1855
Subfamily Lamprochernetinae Beier, 1932
Genus Lasiochernes Beier, 1932
Type species:
Chelifer (Trachychernes) pilosus Ellingsen, 1910, by original designation.
The key requires no morphometric measurements and is based on males only:
Trichobothrium isb situated midway between ist and it, or slightly closer to ist................................................................................2
Trichobothrium isb much closer to it than to ist; greatly thickened pedipalpal femur which is densely pilose on prolateral face................................................................................L. jonicus (Beier, 1929) [Greece]
Pedipalpal femur and trochanter with numerous short setae on prolateral face................................................................................L. punctiger Beier, 1959 [Democratic Republic of Congo]
Pedipalpal femur and trochanter without numerous short setae on prolateral face................................................................................3
Pedipalpal femur with at least some long setae................................................................................4
Pedipalpal femur without long setae................................................................................12
Pedipalpal patella with some longer setae................................................................................5
Pedipalpal patella without longer setae on prolateral face................................................................................11
All pedipalpal segments with numerous setae, uniformly arranged................................................................................L. turcicus Beier, 1949 [Turkey and Israel]
At least one pedipalpal segment without numerous setae................................................................................6
Chelal hand with at least some long setae................................................................................7
Chelal hand without long setae................................................................................10
Chelal hand with numerous long setae................................................................................8
Chelal hand with long setae restricted to base of chelal hand................................................................................9
Pedipalpal segments more densely setose................................................................................L. siculus Beier, 1961 [Italy]
Pedipalpal segments less densely setose................................................................................L. villosus Beier, 1957 [Turkey]
Chelal hand with long setae restricted to prolateral base of chelal hand.......................................................L. anatolicus Beier, 1963 [Turkey]
Chelal hand with long setae restricted to retrolateral base of chelal hand............................................................L. graecus Beier, 1963 [Albania, Greece]
Chelal hand with moderately dense setation...........................................................L. pilosus (Ellingsen, 1910) [Central and Western Europe]
Chelal hand with sparse setation..............................L. congicus Beier, 1959 [Democratic Republic of Congo]
Pedipalpal femur with long setae on basal half of prolateral face.........................................................L. marinae sp. nov. [Čočina jama, Croatia]
Pedipalpal femur without long setae on basal half of prolateral face....................................................L. pavlekae sp. nov. [Croatia]
Chelal hand with long setae on prolateral face................................................................................L. cretonatus Henderickx 1998 [Crete]
Chelal hand without long setae on prolateral face................................................................................L. jalzici sp. nov. [Vilina špilja, Croatia]
Lasiochernes marinae Hlebec & Harvey, sp. nov.
Zoobank registration: http://zoobank.org/NomenclaturalActs/99902BA6-99D4-45F3-B30B-F8287106EA63
(Figs 4–6)
Type material:
Holotype ♂, CROATIA: Dubrovnik-Neretva County: Metković, Nova Sela, Čočina jama, 43°06ʹ40.3″N, 17°34ʹ00.5″E, 215 m a.s.l. (cave entrance), 8 October 2022, B. Jalžić (CNHM408). Paratype: 1♀, data as holotype except 10 February 2019 (CNHM491).
Diagnosis:
Lasiochernes marinae differs from other Lasiochernes species by the presence of long setae (N < 30) situated basally on the prolateral face of the male femur (Figs 4C, 5A), and the slightly rounded posterolateral margin at coxa IV, especially in females (Fig. 4C, D).
Lasiochernes marinae sp. nov.. A, holotype male, habitus (CNHM779), dorsal. B, paratype female, habitus (CNHM491), dorsal. C, holotype male, cephalothorax, ventral. D, paratype female, cephalothorax, ventral. Scale lines = 1 mm (A, B), 0.5 mm (C, D).
Lasiochernes marinae sp. nov.. A, holotype male (CNHM779), right pedipalp, dorsal. B, paratype female (CNHM491), right pedipalp, dorsal. C, holotype male, left chela, retrolateral. D, paratype female, left chela, retrolateral. Scale lines = 0.25 mm.
Holotype:
CNHM408 (DNA voucher CROBD205; GenBank ON841932) served as the reference sequence. Paratype: CNHM491 (DNA voucher CROBD290; GenBank ON842134) shared COI haplotype.
See Supporting Information, Appendix S6 for a list of diagnostic nucleotides in the barcode fragment of the COI gene. Uncorrected pairwise genetic distances (p-distances) between Lasiochernes marinae sp. nov. and other Lasiochernes species are listed in the Supporting Information, Appendix S9. Positions refer to the alignment available in the Supporting Information, Appendix S10.
Description (adults)
Colour:
Pedipalps, carapace and coxae deep red-brown; tergites and sternites reddish; legs yellow-brown (Fig. 4).
Pedipalp (Fig. 5):
All pedipalpal surfaces granulate; segments setose, femur more setose in male; most setae large, curved and dentate; prolateral face of femur of male with long setae, situated in basal half, basal setae straight, medial setae slightly curved (Fig. 5A); patella with several dorsal lyrifissures (Fig 5A, B); all segments robust, slightly longer and thinner in female; trochanter 1.62 (♂), 1.76 (♀), femur 2.46 (♂), 2.61 (♀), patella 2.34 (♂), 2.53 (♀), chela (with pedicel) 2.80 (♂), 2.94 (♀), chela (without pedicel) 2.69 (♂), 2.70 (♀), hand (without pedicel) 1.26 (♂), 1.32 (♀) × longer than broad, movable finger (without pedicel) 0.87 (♂), 0.98 (♀) × longer than hand. Fixed chelal finger with eight trichobothria, movable chelal finger with four trichobothria (Fig. 5A, B): eb and esb situated basally, est situated slightly closer to esb than to et, ib and ist situated subbasally, isb situated opposite est (♂) or slightly posterior to est (♀) and midway between ist and it (♂) or closer to ist than to it, and it situated subdistally; t situated subdistally, st situated slightly closer to t than to sb, and sb situated much closer to b than to st (Fig. 5C, D). Venom apparatus only present in movable chelal finger, venom duct long, terminating in nodus ramosus basal to t. Pseudotactile setae absent. Chelal fingers not gaping; teeth slightly retrorse, juxtadentate, basal teeth more rounded; fixed finger with 42 (♂), 40 (♀) teeth, plus 10 (♂), 10 (♀) retrolateral accessory teeth and 5 (♂), 6 (♀) distal prolateral accessory teeth; movable finger with 45 (♂), 42 (♀) teeth, plus 9 (♂), 8 (♀) retrolateral accessory teeth and 10 (♂), 4 (♀) distal prolateral accessory teeth; sense-spots: fixed chelal finger with 12 (♂), or 11 (♀) retrolateral and 10 (♂) or 2 (♀) prolateral spots, movable chelal finger with 5 (♂), 8 (♀) retrolateral and 0 (♂, ♀) prolateral spots.
Chelicera:
With five setae on hand and one subdistal seta (gs) on movable finger; setae es, bs and sbs lightly dentate, ls and is long and acuminate (Fig. 6A, B); galea well-developed, long and thin with two distal and three medial rami (♂) and two distal and four medial rami (♀) (Fig. 6D, E); with two dorsal lyrifissures and one ventral lyrifissure; rallum composed of three blades that are dentate on anterior face (Fig. 6C); serrula exterior with 21 (♂) and 18 (♀) blades; lamina exterior present.
Lasiochernes marinae sp. nov.. A, holotype male (CNHM779), left chelicera, dorsal. B, paratype female (CNHM491), left chelicera, dorsal. C, paratype female, left rallum, lateral. D, holotype male, left galea, dorsal. E, paratype female, left galea, dorsal. F, holotype male, genital opercula (sternites II and III). G, holotype male, left leg IV, retrolateral. H, paratype female, genital opercula (sternites II and III). I, paratype female, spermathecae. Scale lines = 0.25 mm (G), 0.1 mm (A, B, F, H, I), 0.05 mm (C, D, E).
Carapace:
Granulate and sub-rectangular; with two transverse furrows, posterior furrow situated slightly closer to posterior margin than to anterior furrow, posterior furrow with small posteromedian furrow (Fig. 4A, B); 1.04 (♂, ♀) × longer than broad; eyes and eye-spot absent; carapace with 47 (♂), 74 (♀) setae including 7 (♂, ♀) setae near anterior margin and 8 (♂), 10 (♀) setae near posterior margin.
Coxal region:
Maxillae rugose, coxae smooth; manducatory process pointed, with 3 apical acuminate setae (♂, ♀), one small sub-oral seta, and 36 (♂), 39 (♀) additional setae. Chaetotaxy of coxae I–IV: ♂: 29: 31: 43: 56; ♀: 27: 37: 40: 92. Coxa IV with slightly rounded posterolateral margin.
Legs:
Femur + patella of leg IV 4.25 (♂), 5.08 (♀) × longer than deep; tarsus IV with long tactile seta located medially, TS ratio = 0.51 (♂), 0.53 (♀); arolium shorter than claws, claws simple (Fig. 6G).
Abdomen:
Tergites I‒II partially divided, other tergites and sternites IV‒XI with broad medial suture line. Tergal chaetotaxy: ♂, 12: 12: 11: 13: 15: 13: 15: 16: 16: 11: 6: 2; ♀, 12: 11: 13: 14: 13: 16: 14: 15: 12: 9: 5: 2; setae lightly dentate. Sternal chaetotaxy: ♂, 44: (2) 20 [3 + 3] (3): (4) 14 (4): 18: 20: 18: 16: 17: 13: 8: 2; ♀, 33: (3) 13 (4): (4) 8 (3):16: 15: 18: 22: 18: 16: 13: 2. Tergites and sternites without tactile setae. Pleural membrane wrinkled plicate, without setae.
Genitalia:
Male with typical chernetid morphology; female with single T-shaped spermatheca, length of each receptaculum slightly shorter than median stalk (Fig. 6F, H, I).
Dimensions (mm):
Male holotype: body length 3.21. Pedipalps: trochanter 0.535/0.330, femur 0.860/0.350, patella 0.820/0.350, chela (with pedicel) 1.370/0.490, chela (without pedicel) 1.320, hand (without pedicel) length 0.620, movable finger length 0.710. Chelicera: movable finger length (♀): 0.250. Carapace 0.940/0.905; leg IV: femur + patella 0.765/0.180, tibia 0.620/0.125, tarsus 0.405/0.100, TS = 0.506.
Dimensions (mm):
Female paratype: body length 3.78. Pedipalps: trochanter 0.590/0.335, femur 0.875/0.335, patella 0.885/0.350, chela (with pedicel) 1.560/0.530, chela (without pedicel) 1.430, hand (without pedicel) length 0.700, movable finger length 0.710. Chelicera: movable finger length (♀): 0.210. Carapace 0.995/0.955. Leg IV: femur + patella 0.890/0.175, tibia 0.445/0.145, tarsus 0.360/0.090, TS = 0.527.
Etymology:
This species is named for the first author’s beloved mother, Marina Hlebec.
Remarks:
Lasiochernes marinae sp. nov. is known only from the type locality, Čočina jama cave, in southern Croatia (for ground plan and longitudinal section see: Jalžić et al. 2013). All specimens were found under stone near the cave entrance. No additional specimens could be found in other caves in the area, despite significant collecting efforts. Only a single haplotype was recorded for all genetic markers in all specimens, indicating low genetic diversity. Note that Čočina jama cave is also the type locality of the woodlice Trichoniscus matulici metkovicensis Buturović, 1955. During subsequent visits to the locality, only two individuals of bat (Hipposideros sp.) were recorded.
Lasiochernes jalzici Hlebec & Harvey, sp. nov.
Zoobank registration: http://zoobank.org/NomenclaturalActs/81756E62-FE06-4915-8E81-13F9F7B0D3BC
(Figs 7–9)
Type material:
Holotype ♂, CROATIA: Dubrovnik-Neretva County, Cavtat, Gruda, Vilina špilja, 42°30ʹ23.8″N, 18°23ʹ28.7″E, 210 m a.s.l. (cave entrance), 14 November 2020, B. Jalžić (CNHM650). Paratype: 1♀, data as holotype except 23 March 2021 (CNHM778).
Diagnosis:
Lasiochernes jalzici differs from other Lasiochernes species by the lack of long and dense setation on the male pedipalp femur (Figs 7A, 8A), and the non-rounded posterolateral margin of coxa IV (Fig. 7C, D).
Lasiochernes jalzici sp. nov.. A, holotype male, habitus (CNHM650), dorsal. B, paratype female, habitus (CNHM778), dorsal. C, holotype male, cephalothorax, ventral. D, paratype female, cephalothorax, ventral. Scale lines = 1 mm (A, B), 0.5 mm (C, D).
Lasiochernes jalzici sp. nov.. A, holotype male (CNHM650), right pedipalp, dorsal. B, paratype female (CNHM778), right pedipalp, dorsal. C, holotype male, left chela, retrolateral. D, paratype female, left chela, retrolateral. Scale lines = 0.25 mm.
Holotype:
CNHM650 (DNA voucher CROBD760; GenBank ON841846) served as the reference sequence. Paratype: CNHM778 (DNA voucher CROBD1019; GenBank ON842122) shared COI haplotype.
See Supporting Information, Appendix S6 for a list of diagnostic nucleotides in the barcode fragment of the COI gene. Uncorrected pairwise genetic distances (p-distances) between Lasiochernes jalzici sp. nov. and other Lasiochernes species are listed in the Supporting Information, Appendix S9. Positions refer to the alignment available in the Supporting Information, Appendix S10.
Description (adults)
Colour:
Pedipalps and anterior part of carapace deep reddish-brown, metazone reddish, coxae and tergites and legs yellow brown (Fig. 7).
Pedipalp (Fig. 8):
All pedipalpal surfaces granulate; segments setose, slightly more densely setose in male, setae denticulate and sparsely distributed on chelal hand (♀) (Fig. 8A, B); patella with several dorsal lyrifissures (Fig. 8A, B); all segments robust but thinner in female; trochanter 1.66 (♂), 1.68 (♀), femur 2.32 (♂), 2.69 (♀), patella 2.05 (♂), 2.50 (♀), chela (with pedicel) 2.78 (♂), 3.10 (♀), chela (without pedicel) 2.60 (♂), 2.88 (♀), hand (without pedicel) 1.31 (♂), 1.50 (♀) × longer than broad, movable finger (without pedicel) 0.97 (♂), 1.03 (♀) × longer than hand. Fixed chelal finger with eight trichobothria, movable chelal finger with four trichobothria (Fig. 8A, B): eb and esb situated basally, est midway between esb and et, ib and ist situated subbasally, isb situated midway between ist and it, it situated subdistally, slightly closer to isb than to fingertip (♂); t situated subdistally, slightly closer to fingertip (♀), st situated midway between sb and t, slightly closer to sb (♀), and sb situated much closer to b than to st (Fig. 8C, D). Venom apparatus developed only in movable chelal finger, venom duct long, terminating in nodus ramosus between t and st. Pseudotactile setae absent. Chelal fingers not gaping; chelal teeth juxtadentate, slightly retrorse, basal teeth more rounded; fixed finger with 39 (♂), 42 (♀) small, largely unsclerotized teeth, plus 10 (♂), 11 (♀) retrolateral accessory teeth and 6 (♂), 5 (♀) distal prolateral accessory teeth; movable finger with 44 (♂), 48 (♀) teeth, plus 11 (♂), 9 (♀) retrolateral accessory teeth and 4 (♂, ♀) distal prolateral accessory teeth; sense-spots arranged: fixed chelal finger with 9 (♂) grouped basally, 12 (♀) retrolateral and 0 (♂), 11 (♀) prolateral of which 10 are grouped at the base of the finger, and movable chelal finger with 4 (♂), 6 (♀) retrolateral and 10 (♂) of which 7 are basal to ib and 3 close to isb, 0 (♀) prolateral.
Chelicera:
With five setae on hand and one subdistal seta (gs) on movable finger; setae es, bs and sbs dentate, ls and is long and acuminate (Fig. 9A); galea well-developed, long, and thick with five terminal rami (♂), three terminal rami and three medial rami (♀) (Fig. 9D, E); with two dorsal lyrifissures and one ventral lyrifissure; rallum consisting of three blades with serrated anterior margins (Fig. 9B, C); serrula exterior with 17 (♂) and 20 (♀) blades; lamina exterior present.
Lasiochernes jalzici sp. nov.. A, paratype female (CNHM778), right chelicera, dorsal. B, holotype male (CNHM650), left rallum, lateral. C, paratype female, left rallum, lateral. D, holotype male, left galea, dorsal. E, paratype female, left galea, dorsal. F, holotype male, genital opercula (sternites II and III). G, holotype male, right leg IV, retrolateral. H, paratype female, genital opercula (sternites II and III). I, paratype female, spermathecae. Scale lines = 0.25 mm (G), 0.1 mm (A, F, H, I), 0.05 mm (B, C, D, E).
Carapace:
Granulate and rectangular; approximately as long as broad; eyes or eyespots absent; two transverse furrows; posterior transverse furrow slightly closer to posterior margin of carapace than anterior furrow; posterior furrow with medial longitudinal, both furrows slightly recurved (Fig. 7A, B); with 73 (♂), 77 (♀) short, clavate setae including 7 (♂, ♀) setae near straight anterior margin and 10 (♂), 12 (♀) setae near posterior margin.
Coxal region:
Maxillae and coxae smooth; manducatory process pointed, with three apical acuminate setae, with one small sub-oral seta on medial edge, and 50 (♂), 56 (♀) additional setae. Chaetotaxy of coxae I–IV: ♂: 20: 21: 30: 61; ♀: 30: 35: 47: 82.
Legs:
Femur + patella of leg IV 4.54 (♂), 4.45 (♀) × longer than deep; tarsus IV with long tactile seta located medially, TS ratio = 0.53 (♂), 0.54 (♀); arolium shorter than claws; claws simple (Fig. 9G).
Abdomen:
Tergite I (♂, ♀) barely divided, II–X (♂, ♀) with medial suture line. Tergites without pale patches. Tergal chaetotaxy: ♂, 12: 11: 13: 17: 14: 15: 16: 15: 13: 9: 11: 2; ♀, 13: 13: 18: 20: 19: 18: 21: 16: 15: 13: 6: 2; setae lightly dentate. Sternal chaetotaxy: ♂, 32: (3) 17 [3 + 3] (4): (4) 14 (4): 22: 22: 21: 21: 20: 17: 10: 2; ♀, 32: (3) 11 (2): (5) 8 (4): 23: 24: 25: 22: 25: 15: 13: 2. Tergites and sternites without tactile setae. Pleural membrane wrinkled plicate, without setae.
Genitalia:
Male with typical chernetid morphology; female with single T-shaped spermatheca, length of each receptaculum much shorter than median stalk (Fig. 9F, H, I).
Dimensions (mm):
Male holotype: body length 3.00. Pedipalps: trochanter 0.530/0.320, femur 0.870/0.375, patella 0.820/0.400, chela (with pedicel) 1.450/0.520, chela (without pedicel) 1.350, hand (without pedicel) length 0.680, movable finger length 0.700. Chelicera: movable finger length (♂): 0.230. Carapace 0.970/0.990. Leg IV: femur + patella 0.795/0.175, tibia 0.645/0.145, tarsus 0.420/0.110, TS = 0.530.
Dimensions (mm):
Female paratype: body length 3.26. Pedipalps: trochanter 0.580/0.345, femur 0.900/0.335, patella 0.900/0.360, chela (with pedicel) 1.610/0.520, chela (without pedicel) 1.500, hand (without pedicel) length 0.780, movable finger length 0.760. Chelicera: movable finger length (♀): 0.220. Carapace 1.010/1.000. Leg IV: femur + patella 0.890/0.200, tibia 0.695/0.130, tarsus 0.480/0.100, TS = 0.540.
Etymology:
This species is named for Branko Jalžić (CBSS, Zagreb, Croatia), in honour of his contributions to cave biology and assistance to the first author.
Remarks:
Lasiochernes jalzici sp. nov. is known only from the type locality, Vilina špilja cave in southern Croatia (for ground plan and longitudinal section see Jalžić et al. 2013). All specimens were found under stones near the end of the cave in the dark zone. The ground is partially covered with clay and speleothem is sporadic. No additional specimens could be found in other caves in the area, despite significant collecting efforts. Only a single haplotype was recorded for all genetic markers in all specimens, indicating low genetic diversity. Note that Vilina špilja cave is also the type locality of the gastropod Virpazaria pageti kleteckii Štamol and Subai, 2012 and the harvestman Cyphophthalmus silhavyi (Kratochvíl, 1938). Also note that no bats have been recorded in the cave.
Lasiochernes pavlekae Hlebec & Harvey, sp. nov.
Zoobank registration: http://zoobank.org/NomenclaturalActs/B1738410-6E08-448A-8BD0-F5CFF08AABD9
(Figs 10–14)
Type material:
Holotype ♂ (CBSSI608), CROATIA: Dubrovnik-Neretva County: Pelješac, Orebić, Čagjina jama cave, 42°58ʹ37.6″N, 17°09ʹ18.7″E, 107 m a.s.l. (cave entrance), 3 November 2016, F. Belak. Paratypes 1♂ (CBSS292A), 4♀♀ (CBSSI292B, CBSSI292C, CBSSI292D, CBSS32), same data except 14 November 2018, F. Šarc; 2♀♀ (CBSSI293, CBSS43), same data except 14 November 2018, P. Visković.
Additional material examined:
1♂ (CNHM470), 42°50ʹ48.8″N, 17°38ʹ12.1″E, 415 m a.s.l. (cave entrance): Pelješac, Ston, Vukasi, Bijelo jezero, 29 March 2019, B. Jalžić; 1♂ (CBSS312A), 42°84ʹ72.7″N, 17°63ʹ64.7″E, 410 m a.s.l. (cave entrance): Pelješac, Ston, Crno jezero, 29 March 2019, A. Kirin; 2♂♂, 1♀ (CBSS37, CBSSI346, WAMT160123), 42°58ʹ45.8″N, 17°17ʹ45.6″E, 275 m a.s.l. (cave entrance): Pelješac, Orebić, Oskorušno, Jama više vodospreme, 23 April 2019, V. Sudar; 1♀ (CBSS344N), same data except 11 October 2022, B. Jalžić; 1♀ (CBSS54), same data except 23 April 2019, L. Ružanović; 1♂ (CNHM657), 42°55ʹ08.4″N, 17°23ʹ12.8″E, 250 m a.s.l. (cave entrance): Pelješac, Trstenik, Čućin Vrh, Prnčeva špilja, 10 November 2020, B. Jalžić; 1♂, 1♀ (CNHM780, CNHM781), same data except 11 October 2022; 3♂♂ (CBSS502A, CBSS502B, WAMT160122), same data except 10 November 2020, M. Pavlek; 1♂ (CBSSI319), 42°97ʹ85.4″N, 17°12ʹ90.9″E, 53 m a.s.l. (cave entrance): Pelješac, Orebić, Kučište, Šimunkovića špilja, 29 April 2019, M. Jagić; 1♂ (CBSS64), 42°52ʹ28.9″N, 17°36ʹ59.4″E, 434 m a.s.l. (cave entrance): Pelješac, Ston, Sparagovići, Vranja jama, 16 November 2018, P. Visković; 1♂, 2♀♀ (CNHM628, CNHM625, CNHM627), 42°40ʹ37.9″N, 18°08ʹ10.3″E, 103 m a.s.l. (cave entrance): Dubrovnik, Rožat, Vilina špilja-Izvor Omble cave, 29 June 2020, B. Jalžić; 1♂ (CBSSI533), 42°73ʹ46.2″N, 16°91ʹ06.9″E, 100 m a.s.l. (cave entrance): Lastovo, Radaš do, Rača, 10 September 2018, M. Lukić; 1♂, 1♀ (CNHM779, CNHM778), 42°45ʹ29.2″N, 17°29ʹ16.8″E, 190 m a.s.l. (cave entrance): Mljet, Babino Polje, Blato, Dočina špilja, 23 February 2021, B. Jalžić; 1♂ (CBSSI423), 42°73ʹ17.8″N, 17°54ʹ24.1″E, 128 m a.s.l. (cave entrance): Mljet, Babino Polje, Jama pod Sv. Spasom, 19 June 2021, A. Ćukušić; 1♂, 1♀ (CBSSI424A, CBSSI424B), 42°76ʹ44.4″N, 17°47ʹ10.7″E, 55 m a.s.l. (cave entrance): Mljet, Babino Polje, Blato, Mala spila, 16 October 2021, M. Pavlek; 1♀ (CNHM513), same data except 28 September 2019, B. Jalžić; 1♂ (CBSSI458), 42°76ʹ13.3″N, 17°43ʹ08.0″E, 265 m a.s.l. (cave entrance): Mljet, Babino Polje, Goveđari, Male ponte, 16 June 2021, A. Ćukušić; 1♀ (CBSSI595A), same data except 14 February 2021, T. Rožman; 1♀ (CBSS436I), 42°45ʹ30.2″N, 17°26ʹ10.0″E, 201 m a.s.l. (cave entrance): Mljet, Babino Polje, Goveđari, Špilja kod Nerezinog dola, 16 October 2021, N. Kuharić; 1♂ (CBSSI453), 42°76ʹ71.1″N, 17°45ʹ36.1″E, 102 m a.s.l. (cave entrance): Mljet, Babino Polje, Blato, Špilja na Strmici, 17 June 2021, H. Cvitanović; 1♀ (CBSSI99A) same data except 20 October 2015, M. Pavlek; 1♀ (CBSSI430), 42°76ʹ46.8″N, 17°47ʹ20.4″E, 79 m a.s.l. (cave entrance): Mljet, Babino Polje, Blato, Velika špilja, 16 June 2021, M. Pavlek; 1♀ (CBSS433) same data except 06 October 2021, I. Čupić; 1♀ (CBSSI534A), 42°75ʹ25.5″N, 17°55ʹ11.6″E, 210 m a.s.l. (cave entrance): Mljet, Babino Polje, Movrica, 15 June 2021, M. Pavlek; 1♂, 1♀ (CBSSI91, CBSSI188), 43°77ʹ42.6″N, 15°34ʹ97.1″E, 15 m a.s.l. (cave entrance): Šibenik-Knin County, Kornati, Gustac Islet, Vjetruša, 18 April 2018, T. Čuković and A. Ćukušić.
Diagnosis:
Lasiochernes pavlekae sp. nov. differs from other Lasiochernes species by the presence of dense setation patterns on all pedipalpal segments (♂, ♀), slightly more setose in males (Figs 10A, B, 11A, B), and a normal posterolateral margin at coxa IV (Fig. 10C, D). Pedipalpal setation is absent in Lasiochernes jalzici sp. nov. (Figs. 7, 8) and pronounced basally on the prolateral face of the male femur in Lasiochernes marinae sp. nov. (Figs 4, 5).
Lasiochernes pavlekae sp. nov.. A, holotype male, habitus (CBSSI608), dorsal. B, paratype female, habitus (CBSSI293), dorsal. C, holotype male, cephalothorax, ventral. D, paratype female, cephalothorax, ventral. Scale lines = 1 mm (A, B), 0.5 mm (C, D).
Lasiochernes pavlekae sp. nov.. A, holotype male (CBSSI608), right pedipalp, dorsal. B, paratype female (CBSSI293), right pedipalp, dorsal. C, holotype male, left chela, retrolateral. D, paratype female, left chela, retrolateral. Scale lines = 0.25 mm.
Paratype:
CBSS32 (DNA voucher CROBD583; GenBank ON841917)served as the reference sequence, because the amplification of the COI gene fragment from the holotype was unsuccessful.
Other sequenced material: CBSSI312A (DNA voucher CBSSI312A; GenBank OQ330000); CBSSI319 (DNA voucher CBSSI319; GenBank OQ330001); CBSSI595A (DNA voucher CBSSI595A; GenBank OQ330002); CBSSI533 (DNA voucher CBSSI533; GenBank OQ330003); CBSSI534A (DNA voucher CBSSI534A; GenBank OQ330004); CBSSI91 (DNA voucher CBSSI91; GenBank OQ330005); CNHM470 (DNA voucher CROBD269; GenBank ON842238); CNHM657 (DNA voucher CROBD767; Gen-Bank ON842072); CBSS37 (DNA voucher CROBD588; GenBank ON842232); CBSS436I (DNA voucher CROBD1194; GenBank ON841949); WAMT160123 (DNA voucher CROBD599; GenBank CROBD599; GenBank ON842212); CBSS54 (DNA voucher CROBD607; GenBank ON842045); CBSS64 (DNA voucher CROBD617; GenBank ON842194); CNHM627 (DNA voucher CROBD670; GenBank ON842271); CNHM628 (DNA voucher CROBD671; GenBank ON842142); and CNHM779 (DNA voucher CROBD1015; GenBank ON842083).
See the Supporting Information, Appendix S6 for a list of diagnostic nucleotides in the barcode fragment of the COI gene. Uncorrected pairwise genetic distances (p-distances) between Lasiochernes pavlekae sp. nov. and other Lasiochernes species are listed in the Supporting Information, Appendix S9. Positions refer to the alignment available in the Supporting Information, Appendix S10.
Description (adults)
Colour:
Pedipalps, carapace and coxae deep red-brown; tergites and sternites brown, darker in female; legs yellow-brown (Fig. 10).
Pedipalp (Figs 11, 13A):
All pedipalpal surfaces granulate; all segments setose, all setae long (Figs 10A, B, 12A, B); patella with several dorsal lyrifissures (Fig. 11A, B); all segments robust, more robust in male; trochanter 1.71 (♂), 1.79 (♀), femur 2.18 (♂), 2.73 (♀), patella 2.13 (♂), 2.12 (♀), chela (with pedicel) 2.82 (♂), 3.00 (♀), chela (without pedicel) 2.66 (♂), 2.76 (♀), hand (without pedicel) 1.22 (♂), 1.39 (♀) × longer than broad, movable finger (without pedicel) 0.90 (♂), 0.97 (♀) × longer than hand. Fixed chelal finger with eight trichobothria, movable chelal finger with four trichobothria (Fig. 11A, B): eb and esb situated basally, est situated midway between esb and et, ib and ist situated subbasally, isb situated slightly posterior to est and closer to ist than to it, and it situated subdistally; t situated subdistally, st situated closer to t than to sb, and sb situated much closer to b than to st (Fig. 11C, D). Venom apparatus only present in movable chelal finger, venom duct long, terminating in nodus ramosus basal to t. Pseudotactile setae absent. Chelal fingers not gaping; teeth slightly retrorse, juxtadentate; fixed finger with 44 (♂), 42 (♀) teeth, plus 4 (♂), 7 (♀) retrolateral accessory teeth and 5 (♂, ♀) distal prolateral accessory teeth; movable finger with 45 (♂), 43 (♀) teeth, plus 9 (♂), 7 (♀) retrolateral accessory teeth and 2 (♂), 6 (♀) distal prolateral accessory teeth; sense-spots arranged: fixed chelal finger with 12 (♂), 15 (♀) retrolateral from which 13 arranged at the base of the finger, 12 (♂), 9 (♀) prolateral, and movable chelal finger with 2 (♂), 4 (♀) retrolateral and 0 (♂, ♀) prolateral.
Chelicera:
With five setae on hand and one subdistal seta (gs) on movable finger; setae es, bs and sbs strongly dentate, ls and is long and acuminate (Figs 12A, B, 13C); galea well-developed, long and thin with three distal and three medial rami (♂, ♀) (Fig. 12D, E); with two dorsal lyrifissures and one ventral lyrifissure; rallum composed of three blades, all three dentate on anterior face (Fig. 12C); serrula exterior with 19 (♂, ♀) blades; lamina exterior present.
Lasiochernes pavlekae sp. nov. A, holotype male (CBSSI608), left chelicera, dorsal. B, paratype female (CBSSI293), left chelicera, dorsal. C, paratype female, left rallum, lateral. D, holotype male, left galea, dorsal. E, paratype female, left galea, dorsal. F, holotype male, genital opercula (sternites II and III). G, holotype male, left leg IV, retrolateral. H, paratype female, genital opercula (sternites II and III). I, paratype female, spermathecae. Scale lines = 0.25 mm (G), 0.1 mm (A, B, F, H, I), 0.05 mm (C, D, E).
Lasiochernes pavlekae sp. nov., male, scanning electron micrographs. A, left pedipalp, dorsal. B, right chela, dorsal. C, chelicera, dorsal. Scale lines = 1 mm (A), 0.5 mm (B), 150 μm (C).
Carapace:
Granulate and sub-rectangular; 1.13 (♂) and 1.07 (♀) × longer than broad; eyes and eye-spot absent; with two transverse furrows, posterior furrow situated slightly closer to posterior margin than to anterior furrow, posterior furrow with small posteromedian furrow (Fig. 10A, B); with 106 (♂), 102 (♀) setae including 7 (♂, ♀) setae near anterior margin and 13 (♂), 11 (♀) setae near posterior margin;
Coxal region:
Maxillae rugose, coxae smooth; manducatory process pointed, with three apical acuminate setae, with one small sub-oral seta on medial edge, and 46 (♂), 49 (♀) additional setae. Chaetotaxy of coxae I–IV: ♂: 28: 38: 41: 65; ♀: 34: 29: 50: 69.
Legs:
femur + patella of leg IV 4.87 (♂), 5.27 (♀) × longer than deep; tarsus IV with long tactile seta located medially, TS ratio = 0.49 (♂), 0.51 (♀); arolium shorter than claws, claws simple (Fig. 12G).
Abdomen:
Tergites I‒II partially divided, other tergites and sternites IV‒XI with broad medial suture line. Tergal chaetotaxy: ♂, 16: 18: 19: 25: 23: 22: 22: 21: 23: 15: 2: 2; ♀, 14: 16: 19: 22: 24: 23: 22: 21: 21: 16: 10: 2; setae lightly dentate. Sternal chaetotaxy: ♂, 46: (4) 28 [4 + 3] (4): (4) 19 (4): 24: 29: 29: 28: 28: 27: 14: 2; ♀, 33: (3) 13 (3): (4) 9 (4): 20: 26: 26: 23: 23: 21: 12: 2. Tergites and sternites without tactile setae. Pleural membrane wrinkled plicate, without setae.
Genitalia:
Male with typical chernetid morphology; female with single T-shaped spermatheca, length of each receptaculum slightly shorter than median stalk (Fig. 12F, H, I).
Dimensions (mm):
Male holotype (CBSSI608), followed by 12 other males in parentheses: body length 3.600 (2.940–4.015). Pedipalps: trochanter 0.575/0.335 (0.490–0.660/0.290–0.390), femur 0.930/0.425 (0.775–1.120/0.335–0.490), patella 0.920/0.430 (0.770–1.050/0.380–0.510), chela (with pedicel) 1.520/0.540 (1.390–1.680/0.460–0.615), chela (without pedicel) 1.440 (1.290–1.540), hand (without pedicel) length 0.660 (0.640–0.800), movable finger length 0.770 (0.660–0.815). Carapace 1.090/0.960 (0.990–1.200/0.820–1.160). Leg IV (holotype): femur + patella 0.950/0.195, tibia 0.770/0.160, tarsus 0.495/0.105, TS = 0.490.
Dimensions (mm):
Female paratype (CBSSI293), followed by 16 other females in parentheses: body length 4.950 (3.500–5.100). Pedipalps: trochanter 0.625/0.350 (0.490–0.690/0.275–0.395), femur 1.010/0.370 (0.775–1.150/0.300–0.415), patella 0.890/0.420 (0.750–1.040/0.315–0.460), chela (with pedicel) 1.650/0.550 (1.330–1.780/0.430–0.600), chela (without pedicel) 1.520 (1.290–1.650), hand (without pedicel) length 0.765 (0.605–0.790), movable finger length 0.790 (0.620–0.860). Carapace: 1.190/1.110 (0.905–1.250/0.920–1.340). Leg IV (paratype): femur + patella 1.080/0.205, tibia 0.845/0.135, tarsus 0.510/0.105, TS = 0.510.
Remarks:
Lasiochernes pavlekae has been recorded in seven caves on the Pelješac Peninsula, Vilina špilja-Izvor Omble cave, as well as 10 additional caves on islands (Mljet Island, Lastovo Island, and Gustac Islet in the Kornati archipelago). All of these locations are within a aerial distance of less than 50 km from the coast. Most caves range from 7 to 114 m in total length, with the exception of Špilja Crno jezero cave, which is 239 m long, and Vilina špilja-Izvor Omble cave, which streches 3050 m in length. The entrances of these caves are situated at elevations between 15 and 434 m above sea level. The majority of them harbour colonies of bats, including maternity colonies of the Mediterranean horseshoe bat (Rhinolophus euryale Blasius, 1853), the greater horseshoe bat [R. ferrumequinum (Schreber, 1774)], Geoffroy’s bat [Myotis emarginatus (E. Geoffroy, 1806)], and the lesser mouse-eared bat (M. blythii Tomes, 1857) (Pavlinić et al. 2010).
Twelve haplotypes were found in the COI gene (p-distances = 0.18–1.09%) and two in 28S (p-distances = 0.36%), but other genes had one haplotype demonstrating that there is low genetic variability within this species despite some variation in the fast-evolving mitochondrial genome.
Etymology:
This species is named for the Dr. Martina Pavlek, a Croatian biospeleologist and taxonomic specialist in subterranean spiders.
Discussion
Integrative taxonomy
In the Dinaric Karst area, pseudoscorpions exhibit high diversification rates, comprising both morphologically distinct and cryptic taxa that can only be identified using molecular markers (Hlebec et al. 2023). Generally, the taxonomy of pseudoscorpions in the region is in a very chaotic state, but first steps are now being undertaken towards the meaningful classification of species and the analyses of their evolutionary histories through the application of DNA sequencing data in taxonomy. This integrative approach, which combines multiple independent lines of evidence (Gąsiorek et al. 2021, Christophoryová et al. 2023), has revealed extraordinary levels of diversity in the region. Our study aims to serve as a pilot study for future taxonomic work on pseudoscorpions in the highly biodiverse Mediterranean Basin.
To assess subfamilial and generic monophyly within Chernetidae, we first conducted traditional key-based identification and subsequently performed phylogenetic analyses using mitochondrial (COI) and three nuclear (28S rRNA, 18S rRNA, and H3) markers. The results of these analyses, combined with our sampling efforts, revealed discrepancies with the current taxonomy at the subfamilial level. However, it is premature to propose alternative, more phylogenetically based, higher-level systematics due to the limited available sequences. To achieve this, broader taxon sampling and a more detailed knowledge of the morphology of both somatic and genitalic structures are required. Many of the genera were described over 60 years ago, and the feasibility of using high-throughput sequencing technologies to extract molecular data from these historical specimens is questionable, particularly because many of them were preserved in lactic acid for morphological examination. Thus, revisiting the type localities to obtain fresh material may be the best solution.
The first molecular characterization of the genus Lasiochernes has confirmed its placement within the subfamily Lamprochernetinae, consistent with its morphology characterized by T-shaped spermathecae. Following key-based identification, we employed three species delineation methods (ABGD, ASAP, and bPTP) that do not require a priori information regarding species morphology, to assess their agreement with phenotype-based delineations. Additionally, we used SEM imaging to examine potential novel morphological traits and conducted a PCA on morphometric data (see the Supporting Information, Appendix S5 for the complete list) to investigate clustering patterns using two populations of Lasiochernes pavlekae sp. nov.. Statistical analyses indicated that slight differences in measured traits can be considered as intraspecific morphological variation. The setation of the male pedipalp emerged as the most reliable character for distinguishing species within the genus Lasiochernes. Our proposed approach can also be applied to the highly diverse pseudoscorpion genera Chthonius C. L. Koch 1843 (19 species in Croatia; WPC 2022) and Neobisium Chamberlin, 1930 (56 species in Croatia; WPC 2022). Both genera contain a high number of uncertain taxa that remain virtually unidentifiable due to inadequate original descriptions. Field sampling focused on type populations, and linking diagnostic sequences and morphology-based species hypotheses should provide a baseline for future taxonomic revisions.
Molecular methods are widely recognized as effective tools for discovering new species, especially in time of biodiversity loss. These methods are particularly valuable in identifying pseudo-cryptic species, where morphological differences are detected upon re-examination prompted by molecular data, as well as cryptic species. In fact, the molecular approach can even serve as the foundation for taxonomic descriptions in such cases. Molecular data are often incorporated into morphological descriptions without attempting to identify diagnostic characters. These characters represent unique traits that are not found in specimens of other closely related species (Goldstein and DeSalle 2011) and enable unambiguous and straightforward delineation. When determining diagnostic nucleotides between closely related taxa, it is crucial to explore the quality of the input data, including the respective alignment, as well as the effects of different processing methods (e.g. alignment strategies) (Jörger and Schrödl 2013).
In our study, we utilized several independent loci and three molecular species delineation approaches that supported the delineation based on morphology. The concordance between morphological and genetic data, along with high interspecific p-distance values, supported the hypotheses that Lasiochernes marinae sp. nov., L. jalzici sp. nov., and L. pavlekae sp. nov. should be treated as distinct species. The ASAP method emerged as the most conservative single-locus delimitation method and, in comparison with tree-based methods, was less prone to MOTU oversplitting (Dellicour and Flot 2018, Magoga et al. 2021).
For the first time, this study demonstrates a correlation between genetic and phenotypic variability within a pseudoscorpion species. The observed genetic homogeneity and lack of haplotype variability between different subterranean populations of the newly discovered species Lasiochernes pavlekae sp. nov. prompted an in-depth investigation of phenotypic variation using morphometrics. However, the results of the Student’s t-test, Mann–Whitney U rank test, and graphical representation of the PC1 and PC2 did not indicate statistically significant differences in the measured traits between specimens from different [‘island’ (Mljet) and ‘mainland’ (Pelješac)] populations (Fig. 14C). These findings suggest that all studied individuals belong to the same species, identified as Lasiochernes pavlekae sp. nov., and that this species may represent an early stage of incipient speciation. This is noteworthy, as incipient speciation is also commonly observed in lineages within the two most speciose families in the area: Chthoniidae and Neobisiidae (Hlebec et al. 2023). Interestingly, in contrast to Chthoniidae and Neobisiidae, where species often exhibit extreme troglomorphic adaptations in the Dinaric Karst (e.g. anophthalmy, reduction of pigment and elongation of appendages), all three species of Lasiochernes described herein do not display major morphological adaptations to a subterranean biology. These species possess normal pigmentation, and the lack of eyes (potentially seen as a subterranean trait) is actually characteristic for the entire genus (Henderickx 1998).
Genetic and morphological variability of Lasiochernes pavlekae sp. nov.. A, MJ haplotype network for L. pavlekae sp. nov. based on the COI gene. Haplotype circles are colour-coded according to localities and circle size is proportional to overall haplotype frequency. Numbers of mutational steps are given as hatch marks. B, sampling localities (colour coding matches insert in Fig. 14A). C, visualization of PCA results performed on morphometric data deriving from the two populations of L. pavlekae sp. nov.. Polygonal shapes indicate the range of variability found in a given population. The points and polygons are coloured according to localities. D, visualization of PCA results performed on morphometric data deriving from male and female specimens of L. pavlekae sp. nov..
Biogeography of Lasiochernes pavlekae
It is a widely known fact that vicariant isolation of animal populations in subterranean habitats leads to diversification. It is not surprising, therefore, that cave-dwelling animals with wide distributions in the Dinaric region, such as the olm, Proteus anguinusLaurenti 1768 (Gorički and Trontelj 2006, Recknagel et al. 2023), cave shrimp (genus Troglocaris) (Zakšek et al. 2009), and the water louse Asellus aquaticus Linnaeus, 1758 (Verovnik et al. 2004), consist of genetically distinct and geographically isolated lineages. In contrast to these well-known textbook examples, our study revealed low degrees of genetic divergence, haplotype sharing, and a lack of statistically significant differences in all analysed traits between populations of Lasiochernes pavlekae sp. nov.. Haplotype sharing was observed between remote islands (Mljet Island and Gustac Islet in Kornati archipelago), which are approximately 210 km apart in aerial distance, as well as between ‘island’ (Mljet), ‘mainland’ (Pelješac), and Vilina špilja-Izvor Omble cave populations.
Following the Messinian salinity crisis (which ended 5.33 million years ago), the Western Balkans experienced alternations of ingressive and regressive sea phases during the Pliocene and Pleistocene, resulting in significant changes to the coastline (Kuhlemann 2003, 2007). One possible explanation for the lack of genetic divergence between the two populations of Lasiochernes pavlekae sp. nov. is that these populations were separated relatively recently, during the Holocene eustatic sea-level rise. Prior to this period, most of the eastern Adriatic islands and the mainland were either completely or partially connected as a single landmass (Delić et al. 2020 and references therein; see Fig. 2 for aerial distances and bathymetry of Adriatic Sea). Active ongoing dispersal (migration sensu population genetics) as a second possibility is unlikely due to the small size of these animals and their specific habitat requirements. A third possibility is passive dispersal, perhaps via phoresy on bats that inhabit these caves. Phoresy refers to an interaction where one organism (the phoront) attaches itself to another organism (the host), possibly in response to environmental changes within the current habitat (Opatova and Št’áhlavský 2018). Pseudoscorpions, which include phoresy in their behavioural repertoire, often exhibit a lack of genetic variation across populations, as previously observed in species such as Larca lata (Hansen, 1885), Allochernes wideri C. L. Koch, 1843 (Ranius and Douwes 2002) and Chernes hahnii C. L. Koch, 1839 (Opatova and Št’áhlavský 2018). Pseudoscorpions have high rates of nucleotide substitution in their mitochondria (Arabi et al. 2012) and phylogenetic structuring should actually be more evident in the genetic data if there was genetic isolation of populations. Only minor variations were observed in this study (e.g. high number of unique haplotypes specific to certain localities and haplotype sharing between distant localities). These findings provide support for the hypothesis of vector-mediated phoresy.
Dispersal mechanisms aside, we suspect that there are several additional populations of Lasiochernes along the coastal areas and islands. However, sampling these populations is challenging due to the difficulty of accessing certain hypogean habitats, such as MSS (Milieu Souterrain Superficiel) habitats. Furthermore, the microhabitat preferences of the species remain unclear, and they are generally rare.
Conservation
Many studies today highlight the significant genetic diversity found in various animal phyla. However, newly discovered species hypotheses remain formally undescribed and are often labeled as putative species, MOTUs (molecular operational taxonomic units), or OTUs (operational taxonomic units). The delay in their formal diagnosis, description, and naming, leads to incomplete taxonomy and hampers conservation efforts (Jörger and Schrödl 2013). Furthermore, these discovered, but not diagnosed taxa, cannot be included in ongoing studies and effectively protected.
Chernetids are rare in caves throughout the Dinaric Karst and greatly outnumbered by species from the families Chthoniidae and Neobisiidae. Two out of three species described here are known from less than four specimens each, despite extensive collecting efforts in caves within the region. Both are single-site endemics and may require conservation management. All caves are listed in the Annex I-Natural Habitat Types of Community Interest whose Conservation requires the Designation of Special Areas of Conservation of the Habitats Directive 92/43/EEC. Additionally, type localities of the described single-site endemics are incorporated in Natura 2000 Data-The European Network of protected sites. Due to their rarity, we discourage further collection of Lasiochernes species in Dinaric Karst, and specimens should be taken from known sites only for scientific purposes.
Conclusion
Further research is needed to investigate the composition, diagnostic features, and monophyly of the families Goniochernetinae and Chernetinae. The findings of this study, which involved morphological and phylogenetic analyses based on four markers, genetic divergence in COI, and species delimitation analyses, support the recognition of Lasiochernes marinae sp. nov., Lasiochernes jalzici sp. nov., and Lasiochernes pavlekae sp. nov. as distinct species, with two of them being single-site endemics and requiring conservation actions. Phylogeographic reconstruction suggests that the current diversity of Lasiochernes pavlekae sp. nov. is primarily the result of multiple dispersal events, likely mediated by bats, rather than being solely shaped by geomorphological history.
To address taxonomic uncertainties surrounding numerous subterranean species in the Dinaric region and to enhance our understanding of species distribution for effective conservation, it is crucial to generate molecular data for as many taxa as possible. These data should be integrated with biological, biogeographical, ecological, and morphological information to achieve a high level of taxonomic resolution. In this sense, our study is a pilot and perhaps a template for future taxonomic approaches in the subterranean biodiversity hotspot of the Balkans.
Acknowledgements
The authors are grateful to all (bio)speleologists from the Croatian Biospeleological Society (Zagreb, Croatia), who participated in the 2nd Biospeleological expedition–Pelješac 2019, and especially Branko Jalžić, Filip Šarc, Paško Visković, Filip Belak, Martina Pavlek, Anđela Ćukušić, Alen Kirin, Vedran Sudar, Hrvoje Cvitanović, Tin Rožman, Nikolina Kuharić, Lea Ružanović, Iva Čupić, Mateja Jagić, Marko Lukić, and Tamara Čuković for supplying specimens, Slavko Polak (Notranjski muzej Postojna) for kindly providing us with the samples of Lasiochernes graecus, Jonas Astrin (Zoological Research Museum Alexander Koenig) for DNA aliquots, and Nikola Tvrtković for discussion regarding bat populations on Croatian islands. Two reviewers and the Editor are thanked for improving the manuscript. This research was funded by Croatian Science Foundation (project DNA barcoding of Croatian faunal biodiversity: IP-2016-06-9988, project leader: M. Kučinić) and D. Hlebec through ESF (DOK-2018-09-1417). Part of this work was funded by the Gorgon Barrow Island Net Conservation Benefits Fund, which is administered by the Department of Biodiversity, Conservation and Attractions and approved by the Minister for Environment after considering advice from the Gorgon Barrow Island Net Conservation Benefits Advisory Board, and we thank Mia Hillyer, Joel Huey, Melissa Danks, and Nerida Wilson of the Western Australian Museum’s Molecular Systematic Unit for some of the sequence data used in this study. The collection of the material was undertaken with permits from the Ministry of Economy and Sustainable Development of the Republic of Croatia (UP/I-612-07/15-48/19, UP/I-612-07/16-48/162, UP/I-612-07/18-48/170, UP/I-612-07/21-48/29). Additional financial support was provided from the DAAD (German Academic Exchange Service, Grant/Award Number: 91809126), and LinnéSys: Systematics Research Fund from the Linnean Society of London and the Systematics Association, awarded to first author.
Conflict of interest
The authors declare no conflicts of interest.
Data availability
The newly generated sequences were deposited in GenBank (accession numbers available in Supporting Information, Appendix S2) and the BOLD database (under the dataset Lasiochernes, DOI: https://doi.org/10.5883/DS-LASIOCH).
Zoobank registration: http://zoobank.org/References/29E3CDDB-9FD8-4586-B3EE-BAEDFC9D10D4
